INDEX

Explanations

differences and options in various scenarios

oai_token-act-pair · gpt-3.5-turbo

questions and phrases indicating uncertainty or conditions

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

andem

-0.63

 Purch

-0.62

acan

-0.62

âĿ

-0.60

affiliated

-0.60

 registers

-0.59

oru

-0.59

ð

-0.58

ector

-0.58

 Featured

-0.58

POSITIVE LOGITS

:(

0.66

dfx

0.65

cule

0.63

maxwell

0.62

ENGTH

0.62

HUD

0.62

EV

0.61

 Lantern

0.60

IVER

0.59

'[

0.58

Activations Density 0.460%

differences and options in various scenarios

questions and phrases indicating uncertainty or conditions

No Comments

No Known Activations